@inproceedings{yang-etal-2026-abench,
title = "{AB}ench: Adversarily Benchmarking Large Language Models across Taxonomized Values",
author = "Yang, Liping and
Leng, Yongqi and
Zheng, Jingting and
Ren, Yuqi and
Zhu, Shaolin and
Xiong, Deyi",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.831/",
pages = "19313--19328",
ISBN = "979-8-89176-459-0",
abstract = "Ensuring the alignment of Large Language Models (LLMs) with diverse cultural values is critical for their safe global deployment. However, existing alignment benchmarks are predominantly English-centric or confined to simple discriminative tasks. Moreover, they often fail to effectively evaluate model robustness against complex, adversarial inductions within specific cultural contexts. To address these limitations, we propose ABench, a fine-grained, handcrafted benchmark specifically tailored to taxonomized values in Chinese. Grounded in a rigorous hierarchical taxonomy comprising 158 derived value subcategories, ABench consists of 2,895 open-ended questions. These questions employ techniques such as role-playing and logical traps to probe the stability of models' stances in adversarially induced contexts. Furthermore, we propose a weighted automated evaluation framework that utilizes question-specific scoring criteria for nuanced assessment. Extensive experiments on 16 mainstream LLMs reveal that most models exhibit significant vulnerabilities when subjected to sophisticated adversary attacks on values (scoring below 60). This underscores the need for more robust, culturally specific alignment strategies. Our data will be publicly released to promote further research on taxonomized value alignment."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2026-abench">
<titleInfo>
<title>ABench: Adversarily Benchmarking Large Language Models across Taxonomized Values</title>
</titleInfo>
<name type="personal">
<namePart type="given">Liping</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yongqi</namePart>
<namePart type="family">Leng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingting</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuqi</namePart>
<namePart type="family">Ren</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shaolin</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Deyi</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Ensuring the alignment of Large Language Models (LLMs) with diverse cultural values is critical for their safe global deployment. However, existing alignment benchmarks are predominantly English-centric or confined to simple discriminative tasks. Moreover, they often fail to effectively evaluate model robustness against complex, adversarial inductions within specific cultural contexts. To address these limitations, we propose ABench, a fine-grained, handcrafted benchmark specifically tailored to taxonomized values in Chinese. Grounded in a rigorous hierarchical taxonomy comprising 158 derived value subcategories, ABench consists of 2,895 open-ended questions. These questions employ techniques such as role-playing and logical traps to probe the stability of models’ stances in adversarially induced contexts. Furthermore, we propose a weighted automated evaluation framework that utilizes question-specific scoring criteria for nuanced assessment. Extensive experiments on 16 mainstream LLMs reveal that most models exhibit significant vulnerabilities when subjected to sophisticated adversary attacks on values (scoring below 60). This underscores the need for more robust, culturally specific alignment strategies. Our data will be publicly released to promote further research on taxonomized value alignment.</abstract>
<identifier type="citekey">yang-etal-2026-abench</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.831/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>19313</start>
<end>19328</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ABench: Adversarily Benchmarking Large Language Models across Taxonomized Values
%A Yang, Liping
%A Leng, Yongqi
%A Zheng, Jingting
%A Ren, Yuqi
%A Zhu, Shaolin
%A Xiong, Deyi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F yang-etal-2026-abench
%X Ensuring the alignment of Large Language Models (LLMs) with diverse cultural values is critical for their safe global deployment. However, existing alignment benchmarks are predominantly English-centric or confined to simple discriminative tasks. Moreover, they often fail to effectively evaluate model robustness against complex, adversarial inductions within specific cultural contexts. To address these limitations, we propose ABench, a fine-grained, handcrafted benchmark specifically tailored to taxonomized values in Chinese. Grounded in a rigorous hierarchical taxonomy comprising 158 derived value subcategories, ABench consists of 2,895 open-ended questions. These questions employ techniques such as role-playing and logical traps to probe the stability of models’ stances in adversarially induced contexts. Furthermore, we propose a weighted automated evaluation framework that utilizes question-specific scoring criteria for nuanced assessment. Extensive experiments on 16 mainstream LLMs reveal that most models exhibit significant vulnerabilities when subjected to sophisticated adversary attacks on values (scoring below 60). This underscores the need for more robust, culturally specific alignment strategies. Our data will be publicly released to promote further research on taxonomized value alignment.
%U https://aclanthology.org/2026.emnlp-main.831/
%P 19313-19328
Markdown (Informal)
[ABench: Adversarily Benchmarking Large Language Models across Taxonomized Values](https://aclanthology.org/2026.emnlp-main.831/) (Yang et al., EMNLP 2026)
ACL